Impairment of arterial endothelial function is an early event in atherosclerosis and correlates with all of the major risk factors for cardiovascular disease (CVD). The most widely employed noninvasive measure of endothelial function involves brachial artery (BA) diameter measurement using ultrasound imaging before and after several minutes of blood flow occlusion (Celermajer et al. (1992) The Lancet, 340: 1111-1115). The change in arterial diameter is a measure of flow-mediated vasodilation (FMD). This peripheral measurement correlates strongly with coronary artery endothelial function, a fact that strongly supports its clinical value. However, the high between-laboratory variability of results and cost of instrumentation render this technique unsuitable for routine clinical use.
Endothelial function is both acutely and chronically affected by lifestyle factors that influence CVD risk (Brunner et al. (2005) J. Hypertens., 23: 233-246). Consequently, measures of endothelial function are useful in monitoring response to medication, dietary changes and exercise regimens. Unfortunately, very little work has focused on determining the clinical value of endothelial function measurements for individual patients or on developing measurement methods suitable for routine or continuous monitoring of endothelial function. There are compelling reasons to believe that knowledge of acute variation in endothelial function in an individual is important. Since NO released by the endothelium is a potent inhibitor of leukocyte and monocyte adhesion to the endothelial cell surface, and since adhesion of these cells is widely believed to be a necessary initiating event in atherogenesis (Deanfield et al. (2005) J. Hypertens., 23: 7-17), it is reasonable to infer that the proportion of time that the endothelium is dysfunctional constitutes an important indicator of disease risk. This is the rationale for the development of techniques that are simple and cheap enough to enable regular or continuous measurement of endothelial function.
The two FDA-approved commercially available systems for measuring endothelial function perform measurements that are based on the flow and pulse pressure in resistance vessels (rather than in conduit arteries). The Endo-PAT2000 system from Itamar Medical analyzes the pulse amplitude in the finger before and after application on endothelial stimulus. While about 46% of the observed changes in pulse amplitude are blocked by NO synthase inhibitors, mechanisms other than those mediated by NO significantly contribute towards the response (Nohria et al. (2006) J Appl Physiol, 101(2): 545-548). This is most probably a consequence of the different mechanisms involved in arterial and arteriolar/microvascular vasodilation. Also, the measurement is made in vessels that experience ischemia and the many non-NO-mediated vasodilatory processes that occur under ischemic conditions. It is clinically preferable to perform measurements on arteries such as the brachial artery, the endothelial response of which is highly correlated with that of the coronary arteries (r=0.78, p<0.001, Takase et al. (1998) Am. J. Cardiol., 82(12): 1535-1539). In addition, a review of close to 2,500 studies found that brachial and coronary artery EF have similar power to predict serious cardiovascular events over a follow-up period of 1-92 months (Lerman and Zeiher (2005) Circulation, 111(3): 363-368). The authors of the review assert that “the similar power of coronary and peripheral endothelial dysfunction to predict cardiovascular events and the observation that the cardiovascular events may occur remotely from the site in which the endothelial dysfunction was detected underscore the systemic nature of endothelial dysfunction and its pivotal role in prediction of cardiovascular events.” It is not currently possible to make such strong statements regarding the significance of microvascular endothelial function.
Two large (>1800 subject) cross-sectional studies found an association between EndoPAT measurements and accepted cardiovascular risk factors (Palmisano et al. (2011) Hypertension, 57(3): 390-396; Schnabel et al. (2011) Circulation: Cardiovascular Imaging, 4(4): 371-380). However, the correlations between EndoPAT and FMD were low in the respective studies: r=0.094 (N=1843), and r=0.19 (N=5000). Correlations decreased further when adjusted for age and sex. Also, some of the results suggest the influence of potentially serious confounding factors. For example, while it is well known that endothelial function tends to decrease with age, older subjects exhibited better endothelial response according to Endo-PAT (Hamburg et al. (2008) Circulation, 117(19): 2467-2474).
A second approved device is the Vendys system developed by Endothelix, Inc. of Houston Tex. This system measures the cutaneous reactive hyperemic response using hand skin temperature measurement during two minutes of brachial artery occlusion and ensuing RH. During occlusion, skin temperature drops in the distal hand. As blood flow is restored, the temperature increases. Studies indicate that the recovery of skin temperature is slowed in subjects having higher Framingham risk scores and other metrics of CVD and CVD risk. Interestingly, substantial temperature changes are also observed in the contralateral hand that experiences no reactive hyperemic episode. This suggests significant neural involvement in the response. For this reason and the results of Wong et al. (2003) J. Appl. Physiol., 95: 504-510 it is reasonable to predict that this response cannot be blocked by NOS inhibitors.
There is no doubt that these systems provide clinical value and can identify patients with pooled cardiovascular risk factors. However, it is not clear that these systems can do this better than paper-based scoring methods such as the Framingham risk in general populations. It is also highly probable that sympathetic nervous activation is a significant confounding factor in endothelial function measurements based on arteriolar and microvascular responses.
Much stronger evidence exists that peripheral artery endothelial function provides more than simply a correlate of CVD risk factors. Few clinicians would disagree with the statement that evaluation of EF in conduit arteries has more proven clinical value.